© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 108735
    Prev
    Next
    INDEX
    Explanations

    Math/code special characters

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     nos
    -0.31
     ours
    -0.28
     mess
    -0.28
     kind
    -0.27
     phase
    -0.27
    为æĪij们
    -0.25
     ripe
    -0.25
    mess
    -0.24
    opped
    -0.24
     revolution
    -0.24
    POSITIVE LOGITS
    ä¸Ńåħ³
    0.25
    anga
    0.25
    è¾ķ
    0.25
    å»ºåĽ½
    0.25
     suất
    0.25
    æķĻ导
    0.24
    è¿Ļä¹Łæĺ¯
    0.24
    ç¾İçļĦ
    0.24
    .lista
    0.24
    ):(
    0.24
    Activations Density 33.068%

    No Known Activations